Autonomy, Slavery, and Companion Animals

236

Between the Species

ABSTRACT I attempt to resolve the question of whether keeping animals as pets is akin to slavery by considering the significance of liberty to human beings and to nonhuman animals. I distinguish between two senses of liberty: preference liberty and autonomous liberty. Preference liberty is the freedom to satisfy the preferences that one in fact has. Autonomous liberty is the ability to satisfy the preferences that one might have regardless of whether one actually has those preferences. Preference liberty has a value for animals, but autonomous liberty is meaningless for them. As the core wrong of slavery is the restriction of autonomous liberty, I conclude that pet-keeping is not akin to slavery, though in practice it is often morally wrong for other reasons.

Heather Kendrick Central Michigan University

Volume 22, Issue 1

Fall 2018

http://digitalcommons.calpoly.edu/bts/ 237 Heather Kendrick

The Polarized Views of Pet-Keeping The moral defensibility of keeping non-human animals (henceforth, “animals”) is a subject that tends to elicit polarized opinions. Among animal rights advocates, it is sometimes regarded as a form of exploitation akin to slavery. Many in the general public, on the other hand, find this stance laughable in light of the supposedly pampered status of most household pets compared with wild animals. “My cat isn’t my slave,” goes the common jocular rejoinder; “I’m his!” This imagined dilemma between pets as pampered or pets as slaves is a false one. Both could be true, or neither, because slavery and cruel treatment are distinct issues.

All too often, it is supposed that – a few bad owners aside – pets have lives of exceptional happiness. Hilary Bok, while recognizing that “on occasion, life with pets can go badly wrong,” confidently asserts that “domesticated animals such as cats and dogs are typically happy living in human households” and “For the most part, the interest of pets and their owners do not diverge in any serious way” (2011, 769). She regards pet-keeping as much less problematic than animal research or farming because “Pet owners’ intentions are both generally benign and focused on nonhuman animals as individuals” (2011, 770). According to Bok, dogs and cats “have much better lives with humans than they would in the wild” (2011, 769), though David DeGrazia points out that comparing the lives of domestic animals to the lives they would have in the wild makes little sense; even a neglected dog is likely better off than a stray simply because dogs are highly domesticated in a way that makes them unsuited for a life in the wild (DeGrazia 2011, 743-44).

Some authors have rightly called into question the idea that pets are generally treated with such a high level of care. In

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 238 Heather Kendrick

sharp contrast to Bok, Charles Danten regards the relationship as disastrous for animals: “I am not making an assertion that happy pets do not exist, but if they do, they are few and far between. Well-being in a dependent relationship is possible only if all of the needs of the dependent animal are met. In the framework of captivity, this condition can be fulfilled only rarely or partially” (2015, 90). Danten’s view, while very likely true of exotic animals, is implausibly pessimistic for highly- domesticated animals such as dogs, whose adaptation to human contact and domestic life has been cultivated over millen- nia. On the other hand, Bok’s utopian view is no less dubious, and Danten provides numerous anecdotes from his career as a veterinarian to illustrate the selfishness that guides many people’s treatment of their pets. Likewise, Bernard Rollin cites frivolous “euthanasia,” irresponsible acquisition and abandon- ment, ignorance of an animal’s needs and nature, and elective surgery such as de-clawing and ear-cropping as examples of the widespread harms inflicted by pet-keepers on the animals under their care (2006, 297-307), and Varner echoes this (2002, 468-70, 473). Stuart Spencer et al. highlight apparent inconsis- tencies between what the public condemns about farming (de- beaking chickens, for instance) and what they tolerate toward pets (cropping dogs’ tails, for instance) (2006, 24). Though their intention seems to be to present a reductio against these objections to industrial farming, it works better as a condemna- tion and call for reform of pet-keeping practices. In light of the widespread disregard for the needs and desires of pets and the low priority their interests are given in conflicts with human members of the household, the general perception of them as too “pampered” to pose real ethical concerns is much too rosy. Clearly pet-keeping is highly problematic in common practice, contrary to the blithe assertion that pets cannot be slaves because they are so pampered.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 239 Heather Kendrick

Granting that pets are not always or often as happy as one might like to think, the question remains: what if they were? Would ending harmful and irresponsible treatment of pets be enough to transform pet-keeping (or companion animal guardianship, as might be a better term for the ideal relationship) into a just institution? Not if it is indeed a form of slavery, because the deepest wrong of slavery is not in its cruelty. Even the most pleasantly-treated human slave suffers a severe violation by being deprived of liberty. The fact that care and concern may be lavished on an animal does not itself prove that the animal is not a slave. Perhaps even the most idealized pet-human relationship is still “benign slavery.” By that term I do not intend to endorse any slavery at all as morally permissible, nor to lessen the essential wrong of slavery. Instead, I merely use it to desig- nate a (probably merely hypothetical) state of slavery in which the enslaved individual is provided with a pleasant life free of significant suffering and want. If pet-keeping is a form of slavery, then no matter how well we treat our pets, the relationship by its nature is morally wrong.

Those who assert that pets live a luxurious life in order to counter the claim that they are slaves have made two errors: first, they have not taken a hard enough look at how pets are actually treated in common practice, and second, they have failed to realize that being treated kindly and being a slave are not mutually exclusive. Those who contend that pet-keeping is a form of slavery have made a different error, one that I intend to focus on henceforth. I will argue that pet-keeping is not comparable to slavery. It may be that the institution ought still to be abolished, because of the frequent harm done to pet animals by irresponsible pet-keepers, and the difficulty of causing the social change needed to end this harm. Any wrongfulness of pet- keeping, however, does not come from a similarity to slavery.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 240 Heather Kendrick

Criticism and Defenses of Pet-Keeping Several authors have criticized our relationship with domestic animals by evoking the specter of slavery. Marjorie Spie- gel’s The Dreaded Comparison is an extended comparison between human slavery and our treatment of animals, though she discusses pets only briefly, with laboratory and farm animals occupying most of her concern. “We might look at the relationship between a dog and his master, just one example of what is sometimes a modern slave/slave-owner relationship . . . . If a dog wishes to do something other than what pleases his master . . . he may be beaten or otherwise punished. All independent actions are thus discouraged” (Spiegel 1996, 41). Spiegel, however, notably refers to the dog-human relationship as sometimes a slave/slave-owner relationship, which recalls Bok’s argument that the relationship can, with the right attitude on the part of the pet’s caretaker, be morally acceptable.

Danten discusses the wrongfulness of pet-keeping more ex- tensively in his work titled (in English) Slaves of Our Affec- tion: The Myth of the Happy Pet. Most of it is about the cruelty inflicted on pets for the sake of selfishness and convenience rather than on the concept of slavery per se, but the appendix contains a provocative quotation from James Stirling detail- ing his observations of slaves in the American South, to which Danten has added words parenthetically to show how it can easily be adapted to describe our relationships with domestic animals. So when Stirling tells us that “the house servant is comparatively well off” and benefits from greater affection from the master than the plantation hand, Danten reads “pet” for “house servant” and “farm animal” for “plantation hand”; in other words, while pets may be subject to less severe mistreatment, they are no less slaves (2015, 255-56). Again, however, the focus is on the cruelties inflicted: Danten believes that

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 241 Heather Kendrick

the life of a pet is only a less severe mistreatment than that of the farm animal, and believes that pets that are not mistreated at all are rare to nonexistent (Danten 2015; 90, 254).

As previously argued, focusing on the suffering or privation inflicted on pets does not entirely address the question, since it is at least theoretically possible that a pet could be free from significant suffering or privation (including boredom, loneli- ness, or other emotional harms) and still be a slave. Leslie Ir- vine offers a typical example of such an argument, stating that “...the perpetuation of our pleasure is not sufficient reason to enslave other animals” (2004, 5). While acknowledging that a welfarist (utilitarian) position would permit pet-keeping if the institution were significantly modified, she endorses a deonto- logical approach drawn from Tom Regan and Gary Francione, stating that animals have inherent value and therefore cannot be treated as property: “If we recognize the intrinsic value of animals’ lives, then it is immoral to keep them for our pleasure, regardless of whether we call them companions or pets” (Irvine 2004; 11, 14). She concludes that although she does not herself regard the animals in her household as property, “...outside of the household, that is exactly their status. I am free to pamper them or ignore them, as long as I am not caught inflicting in- tentional cruelty” (2004, 14). Francione has made a very simi- lar argument. The institution of “pet ownership,” according to Francione, violates animals’ right not to be treated as things, because it considers pets to be property. “You may treat your animal companion as a member of your family and accord her or him inherent value or the basic right not to be treated as your resource. But your treatment of your animal really means that you regard your animal property as having more than market value; should you change your mind . . . your decision will be protected by the law” (Francione 2007, 169).

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 242 Heather Kendrick

While Bok’s view of pets is dubiously rosy, her reply to Francione (and, by extension, Irvine) is strong. The fact that the law currently treats pets in a certain way does not mean that it always must or that we cannot do better then the law requires in the meantime. “The problem with our current system, is . . . that it gives us rights over nonhuman animals that we arguably ought not to have,” according to Bok. “Fortunately, this system does not force us to exercise those rights. We can, and should, treat our pets in just the same way that we would if the laws governing nonhuman animals are exactly as they should be” (2011, 776). That society regards pets – and all animals – as mere property with no legal rights is true and deplorable, but the question of whether pet-keeping is intrinsically and necessarily wrong remains.

Irvine regards the relationship between human and pet as necessarily exploitative, but this seems to rest on the assumption that “keeping animals for our pleasure” must mean “keeping them for our pleasure without regard to their own interests.” That is the essence of exploitation, but it need not be the case in our relationships with companion animals. Just as I can maintain friendships because they give me pleasure without necessarily treating my friends as mere means for my pleasure, I can take pleasure in relationships with animals without necessarily treating them as mere means. Friendships are (or should be) mutually beneficial, just as the defenders of pet-keeping would argue our relationships with animals can be.

Kristien Hens explicitly rejects the idea that companion animals, particularly dogs, are slaves on the grounds that the relationship would not be so successful if owners treated them as slaves: “Repressive training techniques destroy the dog-human relationship rather than build it... It is indeed questionable

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 243 Heather Kendrick

whether the relationship dog-human would have been so successful if they were merely man’s tools” (2009, 6). Neverthe- less, the challenge is not so easily dismissed. Training techniques can be kind, the owner can be loving and caring, and the animals may benefit in many ways from the relationship, but pets are still under our control. Even if they were granted legal protections that distinguish them from property, they would still not be free agents. What account, if any, can be offered for the legitimacy of such a relationship?

Discussions of the pet-human relationship by authors that accept the moral permissibility of such a relationship have often focused on what duties one has to one’s pets, taking as a given that such relationships are (or can be) morally permissible (Burgess-Jackson 1998, Varner 2002). Among those that address the question of moral permissibility itself, there are three primary approaches. The first is to argue that the captivity of domestic animals is for their own good because we have molded them into creatures who are not capable of living “in the wild.” Gary E. Varner writes, “On any plausible ethical theory, the keeping of pets who meet the conditions for being companion animals and domesticated partners is almost surely going to be permissible. At a base minimum, dogs to a significant degree need to live among humans in order to live well...” (2002, 464; cf. Bok 2011, 777). But this fails to note the difference between “we need to keep these dogs for their own good” and “we should continue to encourage the institution of pet-keeping.” A critic of pet-keeping such as Francione or Ir- vine will argue that we must keep the domestic animals who currently exist but ensure no more of them are bred. “As much as I enjoy living with dogs,” Francione writes, “were there only two dogs remaining in the world, I would not be in favor of breeding them” (2007, 170). The goal ought to be to phase out

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 244 Heather Kendrick

the institution, with the presence of pets in our lives prior to that time being a necessary evil. Bok dismisses this as a possibility without argument (2011, 777), but it is not clearly absurd.

The second approach is to say that pet-keeping is justified so long as the animal derives benefit from it. Varner says that “the keeping of companion animals and domesticated partners can be justified to the extent that both keeper and pet genuinely benefit from the relationship” (2002, 465). Those holding the position that pet-keeping is slavery are unlikely to find this argument convincing. It is reminiscent of Aristotle, who stated that “domestic animals are by nature better than wild ones, and it is better for all of them to be ruled by human beings, since in this way they secure their preservation” (2016, 627), but Aristotle believes the same of the allegedly less-rational human beings whom he construed as “natural slaves.” Just as a non-rational animal is better off being ruled by a rational human being, so it is that “slavery is both advantageous and just” for these natural slaves (Aristotle 2016, 627). This comparison, though offered in defense of human slavery, now has the unintended effect of casting suspicion on arguments that justify control over animals on the basis of the benefits they are alleged to receive. Spiegel regards them as bad-faith attempts to obscure the actual misery of domestic animals, as in the case of a factory farm worker who claims the chickens are “happy” because they are protected from predators and receive regular food, despite the fact that she can plainly see that they live in miserable conditions (1996, 75-76). No doubt such claims that domestic animals are happy are often unfounded and serve only as rationalizations for their use, but let us continue to consider the theoretical possibility that a slave may truly be better off, in the sense of having needs and desires met, by entering into slavery. Such benign slavery would still be slavery.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 245 Heather Kendrick

The third approach is to characterize the relationship as a sort of contract. Rollin takes such an approach: “One may choose to see the human relationship to the dog as involving something like a social contract, in which animals give up their free, wild, pack nature to live in human society in return for care, leadership, and food” (2006, 289). Rollin states that “Ac- cording to some ethologists... humans have actually developed the dog into a creature whose natural pack structure has been integrated into human society, with the human master playing the traditional role of pack leader. It is hard to imagine a more vivid and pervasive example of a social contract...” (2006, 290). He claims that this holds for other domestic animals also (2006, 290).

Clare Palmer offers a convincing criticism of the idea of a domestic animal contract with human beings. Palmer identifies four components of a social contract: a transition from one state to another (e.g. from a state of nature to a state of culture), limitations on some freedoms, free and equal individuals who understand and consent to the contract, and benefits to all par- ties to the contract (1997, 414). Palmer concludes that the relationship with domestic animals meets the first two criteria but the last two are problematic.

The third criterion is a problem because animals cannot understand the contract, and there is unequal power between the contractors (Palmer 1997, 417). She considers different models of the social contract that have emerged to resolve the apparent lack of a literal historical contract (and the fact that the existence of such contract may not in any case be binding on the descendants of the original contractors). On the tacit consent model, certain actions are interpreted as tacit consent to the social contract – for instance, willingly deriving benefits from

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 246 Heather Kendrick

the contract. On the hypothetical consent model, one looks at whether someone “would have consented” to the social contract. Neither of these models applies well to animals, because the fact that domestication has given animals no viable option other than the current state of affairs “makes a mockery of the idea of either tacit or hypothetical consent” (1997, 421). Palmer concludes that “key aspects of the social contract and the domesticated animal contract are fundamentally different: the animal contract could not be said to be created by equal individuals, it is not clearly advantageous to all animals and it is dependent either on a controversial idea of tacit consent or on an extremely abstract kind of hypothetical consent” (1997, 421-422).

None of the three approaches adequately respond to the concern that pet-keeping is a form of slavery. In order to determine whether this concern is warranted, we must inquire into the nature of slavery and its wrongfulness for human beings, and then see how or whether that transfers to animals.

Liberty, Autonomy, and Slavery Slavery clearly involves a limitation of freedom. But the term “freedom” encompasses several different concepts, so it is important to determine which senses of freedom are in- volved. First, one might speak of freedom in regard to the agent’s internal capacities – in other words, freedom of will. Second, one might speak of freedom in terms of how the agent is treated by others or affected by outside forces (for instance, whether restraints are placed on the agent by others). It is important to avoid conflating these two senses of freedom, and also important to understand their relationship with each other. For the sake of precision, various terms have been introduced for the varieties of freedom.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 247 Heather Kendrick

DeGrazia distinguishes between liberty and autonomy. He states that “Liberty or freedom of action is the absence of external constraints – such as prison walls or coercive measures – that impede one from doing what one wants,” (1996, 204) whereas autonomy, following Gerald Dworkin, is “a second- order capacity to reflect critically upon one’s first-order preferences and desires, and the ability either to identify with these or to change them in light of higher-order preferences and values” (Dworkin, quoted in DeGrazia 1996, 205). The tempta- tion might be to regard liberty, so defined, as a more basic and bare-bones variety of freedom, and autonomy as a richer and more robust freedom. They are not precisely paired concepts, however. DeGrazia’s definition of liberty makes it freedom in the external sense (absence of limitations), whereas autonomy refers to freedom in the internal sense (freedom of will).

Alasdair Cochrane uses the terms “liberty” and “freedom” interchangeably, but distinguishes between three senses of liberty. Negative liberty “refers to being free from interference and constraints,” positive liberty “refers to an individual’s ability to control his or her own life: to self-govern and to self- rule,” and republican liberty “refers to the absence of domina- tion” (Cochrane 2009, 663). Cochrane states that “whichever conception is adopted, most people consider freedom to be an absolutely fundamental interest of human beings,” but his own assumption that human slavery always violates an interest in liberty (2009, 665) paired with his admission that “benevolent” slavery may be compatible with negative liberty (2009, 663) suggests that he adopts a positive or republican conception of “liberty.” Cochrane, like DeGrazia, uses Dworkin’s definition of autonomy, interpreting it as “the capacity to frame, revise, and pursue one’s own conception of the good” (2009, 665).

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 248 Heather Kendrick

While I will adopt “liberty” to refer to an external variety of freedom and “autonomy” to refer to an internal variety, I want to introduce further distinctions and clarifications on both sides of the fence. In the domain of free will, we should distinguish between autonomy in DeGrazia’s sense and what might be called freedom of choice. Various authors have used different terms for these ideas while retaining the essential distinction. Kantians refer to autonomy (in Kant’s terms, Wille) as opposed to the power of choice (in Kant’s terms, Willkür). Others have characterized both of these concepts as varieties of autonomy. Regan refers to the Kantian sense of autonomy, which requires the ability to deliberate and reflectively evaluate the merits of different actions, and contrasts it with “preference autonomy,” which only requires that one have “preferences and have the ability to initiate action with a view to satisfying them” (2004, 84-85). Similarly, Steven Wise distinguishes between “full autonomy” (which he identifies with the Kantian conception) and “practical autonomy,” which merely requires the ability to choose appropriate actions with an aim to satisfying desires (2002, 30-32). Andrew Sneddon adopts the terms “deep autonomy” to mean having a life-plan, which “requires reflection, foresight, self-assessment, sensitivity to values that might structure a life, [and] knowledge of the kinds of life one [might] pursue,” and “shallow autonomy” to mean uncoerced choice (2001, 107). For the sake of simplicity, and to avoid confusion, I will reserve “autonomy” to refer to the reflective, end-setting variety (Kantian, full, or deep autonomy), and will use “freedom of choice” for the other capacity (preference, practical, or shallow autonomy).

The term “liberty” conceals another useful distinction. Con- sider, again, DeGrazia’s definition of liberty: “the absence of external constraints – such as prison walls or coercive mea-

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 249 Heather Kendrick

sures – that impede one from doing what one wants.” Cochrane correctly notes that it is possible to imagine a slave who is not impeded from doing what she wants, especially if “what she wants” has been shaped in certain ways (2009, 666). The impediment moves from actual to hypothetical. The slave in our thought experiment is not prevented from doing what she wants, but what she might have wanted. My argument hangs on the significance of this distinction. I will therefore revise De- Grazia’s definition of liberty to distinguish between two kinds of liberty. The first I will call “preference liberty”: the absence of external constraints that impede one from doing what one (in fact) wants. The second I will call “autonomous liberty”: the absence of external constraints that impede one from doing what one might have wanted. Preference liberty is valuable because it permits the exercise of choice, which allows one to pursue (or attempt to pursue) one’s goals. Any creature with preferences and the ability to exercise choice will benefit from preference liberty. Autonomous liberty is, as the name suggests, only relevant to beings possessing autonomy, because it is the ability to reflect on and revise one’s ends that makes such a counterfactual meaningful.

It is generally assumed that normal, adult human beings possess autonomy as well as freedom of choice, but nonhuman animals (at least of the kinds we normally keep as pets) possess only the second of these. “It is highly unlikely that any animal is autonomous in the Kantian sense,” according to Re- gan (2004, 84). Our understanding of animals’ cognitive abili- ties has advanced considerably since Regan first formulated his theory, and they are now credited with much more intelligence than in the past, including remarkable problem-solving abili- ties, in some cases, self-consciousness. Nevertheless, Chris- tine Korsgaard, in her influential Kantian account of duties to

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 250 Heather Kendrick

animals, regards animals as non-rational because they are not aware of the grounds of their actions, a necessary condition for autonomy. She acknowledges that animals exhibit intelligence and may even possess a kind of self-consciousness, but main- tains that autonomy is distinct from these and that, as far as we know, only human beings possess it (Korsgaard 2011, 101-3).

James Rocha takes exception to Korsgaard and makes a case that animals may be minimally rational from a Kantian per- spective, sufficient to warrant the precaution of treating them with a “minimal amount of respect,” though he admits that “animals are still likely to turn out to have limited rational- ity” (2015, 327). Most relevantly, he contends that engaging in play is proof of animals’ capacity for free end-setting: “play involves setting fun for its own sake as an end” (2015, 319). I am skeptical about this account; play is undoubtedly fun, but is the animal really “setting fun as an end” or simply doing something naturally satisfying? But even if play could be described as free end-setting, this falls well short of the idea of autonomy as requiring “reflection, foresight, self-assessment, sensitivity to values that might structure a life, knowledge of the kinds of life one [might] pursue,” etc.) (Sneddon 2001, 107).

Whether any animals possess autonomy is an empirical question. It may be that a few species do have it, and then my argument will not apply to those species. It is likely, however, that even if autonomy does exist in animals, it will be limited to a few species that are not commonly kept as pets. If, as I am supposing, pet animals are not autonomous, only preference liberty will be important to them; autonomous liberty will be irrelevant. The following story will illustrate this distinction.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 251 Heather Kendrick

Imagine that a group of rabbits lives in the middle of a large forest. The range of these rabbits is quite small; they never wander outside a certain clearing in the middle of the forest, not because of any artificial pressure but simply because wan- dering far from their home burrow is not in their nature. Then, imagine that, for some reason or another (perhaps the rabbits are an endangered species and they need protection from human interlopers) a fence is built around the perimeter of the forest. If the rabbits were to wander to the edge of the forest, far from their home territory, they would find themselves con- fined. But they never do wander so far. They do not have any desire to cross it, and never will, so it does not pose an impediment to the pursuit of their desires. The fence does not infringe on the rabbits’ preference liberty.

Now imagine instead that there is a philosopher notorious for never leaving the vicinity of his home town. He has never had a desire to wander far, and in fact reacts very negatively to any suggestion that he ought to travel farther. Then imagine that we set up checkpoints on all the roads that he would take if he ever were to leave town, with guards instructed never to let the philosopher through. That this seems to be entirely different from the situation with the rabbits presents a puzzle. The philosopher, like the rabbits, is not prevented from doing as he wishes, since he does not wish to leave town. So long as he stays this course, he will never bump up against the wall we have placed around him. It seems that we have not restricted his preference liberty either.

One might say, “The difference is that we cannot be certain the philosopher will never try to leave. Perhaps he will become more open-minded about travel as time goes by.” Then let us remove the issue of certainty, and say that we are looking back

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 252 Heather Kendrick

at the philosopher’s life after his death, and it turns out that he indeed never did try to leave town. What harm, what foul, was committed by having the checkpoints set up that he never tried to cross?

The reply then comes, “But if he had tried to leave he would have been prevented.” This concern seems to better address the wrongness of our treatment of the philosopher. What happens if we say the same of the rabbits? “Yes, the fence was far outside the rabbits’ range, but if they had tried to roam much further than normal, they would have been stopped by it.” Though strictly true, it does not seem to have the same moral character. Counterfactuals do not seem as relevant to their situation.

The root of this difference is autonomy. An autonomous being is capable of choosing his or her own ends, and full exercise of autonomy requires that there be viable options. Even if the philosopher fully endorses a homebody existence for himself, for the endorsement to be truly meaningful, the “what ifs” must be open. It must be the case that if he had chosen to be a traveler, he would have been able to do so. This counterfactual is significant for an autonomous being. For a non-autonomous creature, it is not. Their deepest ends are not reflected upon or chosen; they simply are. What matters to them is not choosing between ends, but avoiding frustration in pursuing the ones that they do have. The rabbits do not need the option to do something they will never want to do. What they need is to be allowed to do the things they do, in fact, want to do.

Cochrane argues that liberty has a merely instrumental value for animals, since it allows them to satisfy their desires (Cochrane 2009, 674; cf. DeGrazia 1996, 269). It is the ability to satisfy those desires, not the liberty itself, that has in-

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 253 Heather Kendrick

trinsic value. Liberty has an intrinsic value only for autonomous creatures such as human beings. This is why, according to Cochrane, it is wrong to deprive a human being of liberty, but is not (always) wrong to so deprive an animal. Using the previously-introduced distinction between preference liberty and autonomous liberty, I would instead put it this way. Prefer- ence liberty has a value for non-autonomous animals; autonomous liberty does not. As the example of the homebody philosopher shows, it is possible to compromise autonomy without infringing on preference liberty. Thus human slavery, however benign, wrongs human beings; by its nature, it restricts autonomous liberty. Pet-keeping is not comparable because autonomous liberty has no meaning for most animals, and because pet-keeping is, in theory, compatible with preference liberty.

Liberty and Animals I have used the term “benign slavery” to refer to a state of slavery in which the enslaved individual is provided with a very pleasant life free of suffering and want. Such slavery could, in theory, be consistent with having a great deal of preference liberty. If the enslaved is not significantly hindered in the pursuit of her desires – because she is happy with her lot, and desires nothing else – then she is like the penned-in philosopher: although other options have been closed for her, they are options she is not interested in pursuing.

One may object that such a state of affairs is inconceivable, because someone will always have a desire for freedom even if all of her more worldly desires are met. I am not convinced this is true. In the case of human beings, cultural conditioning, a Brave New World situation, could conceivably result in people who do not yearn for freedom at all. Even if it is true of human beings, however, it is likely not true of animals. An animal may

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 254 Heather Kendrick

wish to escape from a cage, but it is because the cage is inter- fering with pursuing some object of her desire, not because she desires freedom as such. As human beings, we resent the clos- ing off of options in our lives even if we did not desire them because it interferes with our autonomy. Non-autonomous animals are unconcerned with such abstract harms. So if the concept of slavery applies at all to animals, benign slavery would be a possibility for them.

I conclude, however, that the concept of slavery does not apply to the kinds of animals that we keep as pets. The essence of slavery is not the restriction of preference liberty; it is the restriction of autonomous liberty. This can be seen from the fact that benign slavery is still a great wrong to human beings. For a human being to be controlled by another harms her even if she does not wish to leave. It harms her because if she had wanted to leave that life, she would not have been able to. It wrongs her in the way that we wrong the homebody philosopher by pen- ning him in to his home town. It violates her autonomy even if the life dictated for her is perfectly in line with the choices she would make for herself. A non-autonomous animal is not subject to such a violation. Pet animals can be, and are, harmed in many ways, and our treatment of them is often abominable. They may be our victims, but they are not slaves.

Even though pets cannot have their autonomy compromised, it is true that their preference liberty is almost always restricted to some degree. Indeed, the very concept of pet-keeping may require it; it is difficult to imagine an animal whose liberty is never restricted at all but who could still be characterized as a pet. When we restrict a pet’s liberty it is ideally for the animal’s own good: keeping a dog from running into the street and being hit by a car, or forcing a cat to receive a vaccination.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 255 Heather Kendrick

Burgess-Jackson (1998, 182-83) argues that paternalism and at- tendant restrictions of liberty are justified on the same grounds that paternalism toward one’s children is justified. There is a significant disanalogy: children will one day fledge, and animal companions will remain forever within our control.

What justifies our continuing to limit pets’ liberty, but not our adult offsprings’, is that the pets will never achieve autonomy, and the children most likely will. Preference liberty has an instrumental value for animals: it is valuable because it allows them to pursue their well-being. If an animals’ well-being can be improved with some restrictions of liberty, then this is justified. For an adult human being, a restriction of liberty for the sake of well-being is typically not justified because of the restriction of autonomy that it also causes.

Further Questions and Conclusion Although the keeping of companion animals is not intrinsically wrong in the way that enslavement of human beings is wrong, there are significant ethical problems with the way it is currently practiced. Many restrictions of liberty imposed on pets are not done for the animal’s own good but only for the pet-keeper’s. They are often treated with a very low regard for their well-being. It remains an open question whether pet- keeping should be abolished, not because it is akin to slavery, but because the state of affairs that renders it largely unethical today is unlikely to change significantly. The morally acceptable pet-human relationship may be practically impossible to achieve in a widespread way.

If pet-keeping is an institution worth perpetuating, then further questions persist. What animals should we take in as pets? Burgess-Jackson regards making wild animals into pets

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 256 Heather Kendrick

as more questionable than domesticated breeds (1998, 162), but an argument for the moral significance of domestication would be required. I am inclined to suggest that the significance of domestication is simply that it facilitates keeping animals with minimal restrictions of preference liberty. A domestic animal’s desires are more easily satisfied in the domestic environment than a wild animal’s. The moral permissibility of keeping a given animal is dependent on whether that species’s needs can be well met in captivity with minimal infringement of preference liberty. This also hints at the further applicability of my argument to zoos.

What about domestication itself? The argument I have pre- sented makes no claim about the process of domestication; instead I have sought only to evaluate the human relationship with existing domestic animals. It could well be argued that we ought not domesticate any more species, especially as the process of domestication must start somewhere, and the earliest generations will still be wild animals and correspondingly ill- suited to the domestic life. For dogs, cats, and other common pets, that ship sailed long ago.

Finally, we must determine how to characterize the pet- human relationship. Hens, after rejecting master-slave as a good model for the ideal pet-human relationship, considers employer-employee, parent-child, and friend-friend, and finds them all wanting, though with elements of truth to each. I have used “pet” and “pet-keeper” throughout this essay in order to remain neutral, especially as I sometimes discussed less than ideal relationships. For an ideal relationship, “companion animal,” as it is now well recognized and avoids the ownership implications of “pet,” is perhaps the best we can do. The idea that an animal is owned in the same way property is owned

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 257 Heather Kendrick

is unjust and should be changed. Some kind of guardianship model would probably better account for the obligations we have to act in our pet’s interests and not just our own.

Those who regard pet-keeping as nothing less than slavery have made a mistake; they have conflated preference liberty and autonomous liberty. But those, especially in the general public, who regard it as plainly morally unproblematic have made a worse mistake. The current practice is a long way from the ideal conditions that make pet-keeping theoretically acceptable.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 258 Heather Kendrick

References Aristotle. 2016. “Politics.” Translated by C.D.C. Reeve. In Readings in Ancient Greek Philosophy, edited by S. Marc Cohen, Patricia Curd, and C.D.C. Reeve.

Bok, Hillary. 2011. “Keeping Pets.” In The Oxford Handbook of Animal Ethics, edited by Tom L. Beauchamp and R.G. Frey. Oxford: Oxford University Press.

Burgess-Jackson, Keith. 1998. “Doing Right by Our Animal Companions.” The Journal of Ethics 2.2: 159-85.

Cochrane, Alasdair. 2009. “Do Animals Have an Interest in Liberty?” Political Studies 57: 660-79.

Danten, Charles. 2015. Slaves of Our Affection: The Myth of the Happy Pet. Translated by Erin Lestrade. CreateSpace Independent Publishing Platform.

DeGrazia, David. 1996. Taking Animals Seriously: Mental Life and Moral Status. Cambridge: Cambridge University Press.

Francione, Gary L. 2000. Introduction to Animal Rights: Your Child or the Dog? Philadelphia: Temple University Press.

Hens, Kristien. 2009. “Ethical Responsibilities Towards Dogs: An Inquiry into the Dog-Human Relationship.” Journal of Agricultural and Environmental Ethics 22: 3-14.

Irvine, Leslie. 2004. “Pampered or Enslaved? The Moral Di- lemmas of Pets.” International Journal of Sociology and Social Policy 24.9: 5-17.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1 259 Heather Kendrick

Korsgaard, Christine. 2011. “Interacting with Animals: a Kan- tian Account.” In The Oxford Handbook of Animal Eth- ics, edited by Tom L. Beauchamp and R.G. Frey. Oxford: Oxford University Press.

Palmer, Clare. 1997. “The Idea of the Domesticated Animal Contract.” Environmental Values 6.4: 411-25.

Regan, Tom. 2004. The Case for Animal Rights. Berkeley: Uni- versity of California Press.

Rocha, James. 2015. “Kantian Respect for Minimally Rational Animals.” Social Theory and Practice 41.2: 309-27.

Rollin, Bernard E. 2006. Animal Rights and Human Morality. 3ed. Amherst, N.Y.: Prometheus Books.

Sneddon, Andrew. 2001. “What’s Wrong with Selling Yourself into Slavery? Paternalism and Deep Autonomy.” Crítica 33.98: 97-121.

Spencer, Stuart, Eddie Decuypere, Stefan Aerts, and Johan de Tavernier. 2006. “History and Ethics of Keeping Pets: Comparison With Farm Animals.” Journal of Agricul- tural and Environmental Ethics 19: 17-25.

Spiegel, Marjorie. 1996. The Dreaded Comparison: Human and Animal Slavery. New York: Mirror Books/I.D.E.A.

Varner, Gary E. 2002. “Pets, Companion Animals, and Do- mesticated Partners.” In Ethics for Everyday, edited by David Benatar, 150-75. New York: McGraw-Hill.

http://digitalcommons.calpoly.edu/bts/ Vol. 22, Issue 1